Your Theory of Electability is Probably Wrong

Ire and acrimony between the left and center-left is bubbling up as we approach the Iowa caucuses on Monday. The Democratic campaigns are all making their own cases on the basis of ideology, theories of change, party unity and electability for why their standardbearer should be the nominee. The ideological and theory of change arguments are well worn and debated every day in clear and consistent ways.

The cases for electability are much murkier, however. Even though supporters of each candidate tout their own theory of electability as self-obviously correct, there is precious little to show that either the moderates or the progressives or the socialists have a strong case.

The reality is that the average head-to-head polling of each of the leading Democratic candidates against Trump has been remarkably consistent for months now. The RealClearPolitics average for Biden, Sanders and Warren against Trump show Biden leading by 5.4%, Sanders by 3.7% and Warren by 1.8% (those numbers for all candidates would look quite rosier if you discarded the IBD/TIPP poll, which appears to be a significant outlier in favor for Trump.) For months the numbers have shown the same thing: Biden leads Sanders by a point or two against Trump, Sanders leads Warren by a point or two, Warren leads other candidates. But all three defeat Trump.  The latest NBC/WSJ poll is pretty standard:

What’s most interesting about this is how much it submarines both the centrist and leftist arguments for electoral viability.

The more centrist Biden camp claims that only he can make both moderates and culturally conservative white working class voters comfortable with ditching Trump for the safety of the tried and true. The more leftist Sanders camp argues that only he can bring out the the disaffected white working class and mobilize a coalition of young and disaffected voters. Warren argues that only she can unite the disparate factions of the Democratic Party while also mounting a populist campaign against Trump that can persuade disaffected independents.

But there’s really little evidence that any of these arguments are true. The head-to-head polling differences between the three leading Democrats are marginal–particularly taking into account name recognition and candidate familiarity among low-information voters, which favors Biden over Sanders over Warren respectively, and which would be erased in a general election.

The theoretical cases for and against each candidate on electability grounds are also dubious.

Biden’s opponents claim that he will sustain damage from Hunter-Burisma situation, and from his own tendency toward both physical and verbal gaffes. But the Burisma issue has been at the top of the headlines for weeks with no apparent damage to his electability numbers against Trump, and his gaffe-prone tendencies have been known for decades. There is no reason to believe they would become more damaging in September than they are in February. Nor, on the positive side, is there any reason to believe that Biden will get more crossover voters from Trump in the fall than he has now.

Sanders’ opponents like to claim that he isn’t vetted and hasn’t sustained attacks from Republicans that will drive down his numbers. But this is utterly unproven: the sting of attacking “socialism” has weakened to almost non-existent as Republicans have cried wolf about it for decades, and as fewer and fewer voters in the electorate are persuaded by Cold War scare rhetoric in the face of rising inequality and basic costs of living. If the centrist wing of the party had real dirt on Sanders they would be using it by now. And besides, the exact same argument was used in 2008 to claim that Barack Obama would be destroyed in a general election over Reverend Wright and other supposed radicalism. It didn’t happen.

But the flipside argument for Sanders is equally unpersuasive. His supporters like to claim that Sanders will do better against Trump than polling suggests because he will turn out large numbers of non-voters that current polling doesn’t reflect. The results of the next few primaries will tell the tale here, but most polling would in fact reflect newly engaged voters. We also didn’t see Sanders blow past his polling marks in 2016 outside of Michigan, and he actually under-performed in polling in several states (which in turn led to a bevy of wild conspiracy theories among his most ardent support base to explain it away.)

Same for Warren. Her opponents like to claim that the attacks over her claiming Native American ancestry, or her positions on Medicare for All, will doom her in a general election. Yet she continues to defeat Trump in general election polling as usual. On the flip side, there is no evidence that uniting the wings of the party, mobilizing the votes of those who desperately want to see the first woman president, and pursuing a populist politics to win independents is succeeding better than either Biden or Sanders.

The boring reality is that the country is more polarized than it has ever been, and becoming more so. The boring reality is that a realignment is taking shape in which the exurban professional class and white working class increasingly vote for their prejudices over their economics but are declining in numbers, while educated suburbanites, young people and people of color rapidly align with the Democratic Party, on behalf of both moderate and leftist candidates depending in large part on the district. The bluer and more urban the districts, the leftist the viable candidates. A hard-charging progressive like Ocasio-Cortez is more aligned with this coalition in the Bronx than an older establishment incumbent like Joe Crowley, while Democrats of left-center-left ideological alignment perform well in the most purple districts. But even a bisexual Medicare-for-All supporting millennial can win in frontline districts.

Neither Sanders nor Biden are likely to win a landslide against Trump. Both are vulnerable in different ways. Neither are they likely to suffer a dramatic defeat like Kerry, much less McGovern. Warren can definitely defeat Trump, but it will likely be close and she could lose. All of the standard electability arguments are unpersuasive.

The actual evidence suggests that Democrats should vote for the candidate they would like to be president, and leave the electability arguments at the door.

Support the Washington Monthly and get a FREE subscription

David Atkins

David Atkins is a writer, activist and research professional living in Santa Barbara. He is a contributor to the Washington Monthly's Political Animal and president of The Pollux Group, a qualitative research firm.